On Kleinrock's Independence Assumption
| Document type: | Bookchapters |
|---|---|
| Peer reviewed: | Yes |
| Full text: | |
| Author(s): | Adrian Popescu, Doru Constantinescu |
| Title: | On Kleinrock's Independence Assumption |
| Book: | Lecture Notes in Computer Science : Next Generation Internet |
| Year: | 2011 |
| Volume: | 5233 |
| Pagination: | 1-13 |
| Editor: | Demetres D. Kouvatsos |
| ISBN: | 978-3-642-02741-3 |
| Publisher: | Springer-Verlag |
| City: | Heidelberg, Germany |
| Organization: | Blekinge Institute of Technology |
| Department: | School of Computing (Sektionen för datavetenskap och kommunikation) School of Computing S-371 79 Karlskrona +46 455 38 50 00 http://www.bth.se/com |
| Authors e-mail: | adrian.popescu@bth.se, Doru.Constantinescu@telenor.com |
| Language: | English |
| Abstract: | The paper is about analyzing the delay performance in a chain of IP routers, with focus on queueing delay and the correlations existing in a tandem queueing system. Leonard Kleinrock independence assumption is particularly considered, based on which the correlations can be ignored, and the effect on delay performance is negligible, under specific assumptions like for instance Poisson arrival processes, Exponential distribution for packet lengths, sufficient traffic mixing and moderate-to-heavy traffic loads. Furthermore, the paper is reporting on results obtained in experiments done at the Blekinge Institute of Technology in Karlskrona, Sweden, on measurements, modeling and analysis of delay in a chain of IP routers. Particular focus is given to validating the Kleinrock independence assumption regarding the effect of correlations in a tandem queueing system. Our results show that this assumption is not valid in our experiments, and this is particularly observed in the end-to-end delay distribution. |
| Subject: | Computer Science\Networks and Communications Telecommunications\Modelling of Bursty Traffics |
| Keywords: | delay performance, IP routers, tandem queueing system, correlations |
| Note: | (LNCS) 5233, SPIN 12440030, LNCS 5233 ISI no. 000290879900001 |
